Work plan

EcTop 5 aims at an understanding of the fundamental biology of ncRNAs across species, from fungi and plants to viruses and mammals. Towards this goal, we will implement complementary teams with synergistic expertise in cell and molecular biology, virology, bioinformatics, biochemistry and modeling to address the following subtopics: A) biogenesis, structure, function and regulation of ncRNAs, and B) mechanisms and dynamics of ncRNAassociated proteins. A 3rd team will make use of basic knowledge gained by teams A and B for subtopic C) development of biomedical ncRNA applications.


Team A: Molecular mechanisms of non-coding RNA biogenesis and action

A1: Chromatin-associated RNAs in epigenetic regulation: Ingrid Grummt, Maïwen Caudron-Herger & Karsten Rippe

We intend to investigate the mechanism by which nuclear chromatin-associated non-coding RNAs (caRNAs) target epigenetic modifications to specific loci in mammalian cells. The project will focus on the mechanism underlying RNA-dependent de novo DNA methylation. The collaboration will EcTop5 Research Application page 5 take advantage of previous work on RNA-directed DNA methylation conducted in the Grummt group who showed that non-coding RNA may form DNA:RNA triplexes with regulatory gene sequences to target chromatin modifiers (Schmitz et al., 2010). The Rippe lab will contribute methods for the identification of caRNAs by genome-wide sequencing (Caudron-Herger et al., 2011) and correlate their activity with other chromatin features (epigenetic modifications, nucleosome positions, transcription factor binding) in mouse embryonic stem cells and differentiated cells (Teif et al., 2012). The complementary expertise of both labs is expected to decipher the mechanisms that propagate specific epigenetic states through cell division, deregulation of this process being associated with cancer and other diseases.


A2: The role of chromatin in the recognition and degradation of cryptic ncRNAs: Anton Meinhart & Tamas Fischer

A large portion of the eukaryotic transcriptome consists of non-protein-coding RNA (ncRNA) transcripts. While coding RNAs are packaged and exported from the nucleus, the majority of ncRNAs are quickly degraded by the nuclear exosome. Defects in this process lead to toxic accumulation of cryptic transcripts and genomic instability. We previously demonstrated that the histone variant H2A.Z and the histone methyl transferase Clr4 are important factors in the recognition and degradation of cryptic transcripts (Zofall et al., 2009), but their exact role is unclear. The aim of this project is to understand the molecular mechanisms behind this process, and to further study the role of chromatin in RNA processing and stability. In vivo studies will be carried out in the Fischer lab, while the biochemical characterization of the involved complexes will be performed in the Meinhart lab. These studies will significantly increase our understanding of genome organization and genomic stability.


A3: Molecular and structural determinants of transitive RNA silencing: Alexis Maizel & Robert Russell;

Transitive RNA silencing is characterized by the conversion of a miRNA-Argonaute (AGO) mediated cleaved target transcript into double-stranded RNA by a RNA-dependent RNA polymerase and the generation of secondary small interfering RNAs, which in turn can silence other genes. Previously, we have shown that the miR390/AGO7 pair triggers transitive silencing in plants (Marin et al., 2010), and this process is critical for proper development. Why and how only some miRNAs reprogram AGO to trigger transitivity is not understood. This project will combine experimental (Maizel) and in silico (Russell) approaches to uncover the molecular and structural determinants controlling transitive RNA silencing. Critical AGO7 residues will be identified by functional analysis of chimera between AGO7 and AGO1 whereas the structural basis will be modeled using available atomic structures of eukaryotic AGO proteins. The results will have a great impact on the understanding of AGO protein specificity across eukaryotes.


A4: Regulation of protein translation by tRNA methylation: Frank Lyko & Georg Stoecklin

RNA modifications have long been considered to modulate RNA activity, but their function is poorly characterized. Recently, the Lyko and Stoecklin groups have collaborated in the characterization of a mouse strain defective in cytosine-C5 tRNA methylation. These mice showed synthetic lethality, with substantially lower steady-state levels of unmethylated tRNAs and reduced rates of overall protein synthesis (Tuorto et al., 2012). To further characterize the role oft RNA methylation, we now propose a detailed analysis of protein translation. Briefly, polysomes will be isolated from various tissues and mouse embryonic fibroblast lines, fractionated by sucrose centrifugation and analyzed by RNA sequencing using Illumina protocols. Furthermore, polysomes will be nuclease digested for the identification of discrete ribosome footprints by RNA-sequencing. Our results will allow us to determine the effect of tRNA methylation on the ribosome occupancy and translation rate of individual mRNAs.


Team B: The role of non-coding RNAs in cellular physiology and disease models

B1: Removal of non-coding telomeric RNAs at chromosome ends by Sen1/Senataxin: Brian Luke & Karsten Rippe

Telomeres get transcribed into non-coding RNA called TERRA. The Luke lab showed that TERRA causes telomere loss and cellular senescence (Maicher et al., 2012). TERRA forms RNA:DNA hybrids at telomeres, which are responsible for these phenotypes. Yeast Sen1 removes RNA:DNA hybrids and prevents genomic instability. The human homolog, Senataxin, performs similar functions. Senataxin mutations lead to a neurological disorder, ataxia with oculomotor apraxia type 2 (AOA2). Patients have telomere shortening and dysfunction. We will test the hypothesis that TERRA DNA:RNA hybrids are responsible for the telomere dysfunction associated with AOA2. We will test this in yeast, where the Luke lab has established the assays to measure levels of TERRA RNA:DNA hybrids, senescence rates and telomere function. We will then use human tissue culture (patient cell lines) to determine if mammalian TERRA is under the control of Senataxin. We will employ microscopy-based readouts that are established for in the Rippe lab (Chung et al., 2011).


B2: Long non-coding RNAs in Mitosis: Sven Diederichs & Sylvia Erhardt

The function of long non-protein-coding RNAs (lncRNA) is mostly unknown although they cover large parts of the human genome. The Diederichs lab has developed an approach to genetically inactivate lncRNAs (Gutschner et al., 2011), and is currently identifying and characterizing the most abundant tumor-associated lncRNAs. In parallel, it became obvious that RNAs (e.g. derived from microsatellites) play important roles during mitosis and at the kinetochore, which is the focus of the Erhardt lab. Now, we will jointly characterize the role of lncRNAs in mitosis using an siRNA library - designed in the Diederichs lab - that targets 630 tumor-induced lncRNAs. After knockdown, these lncRNAs will be scrutinized by live cell imaging for differences in cell cycle progression, aberrations in mitosis or at the kinetochore. The cellular and molecular function of lncRNAs affecting mitosis will then be studied in detail by depletion and localization studies. RNA affinity purification will identify interacting proteins. Taken together, we aim to unravel the role of cancer-associated lncRNAs during mitosis and for kinetochore function.


B3: Identification of non-coding RNAs controlling iron homeostasis: Martina Muckenthaler & Matthias Hentze

Iron homeostasis must be maintained to prevent frequent disorders of iron overload and iron deficiency. We recently identified the liver-specific miR-122 as a critical regulator of systemic iron homeostasis (Castoldi et al., 2011). In the proposed project we now aim to comprehensively analyze differentially expressed hepatic non-coding RNAs with altered expression levels in mouse models of genetic iron overload [Hfe-/- and Fpn(C326S) mice] by applying genome-wide RNA sequencing. Non-coding RNAs that show altered expression will first be classified (e.g. miRNAs, linc RNAs) and analyzed bioinformatically to unravel complementary sequences within genes involved in the regulation of iron metabolism. Iron-related parameters will be studied for selected non-coding RNAs following their knock-down or overexpression in murine primary hepatocytes and wild-type mice (in cooperation with Dirk Grimm). We expect these findings to have implications for iron-related disorders.


B4: Hepatic miR-122 as a key regulator of iron-dependent physiology and pathology in vivo: Dirk Grimm & Martina Muckenthaler

Mi(cro)RNAs are an important class of ncRNAs critically involved in controlling gene expression in human physiology and pathology. Our aim is to study the basic biology and therapeutic relevance of miRNAs in vivo, using hepatic miR-122 as a prototype due to its vital role in cholesterol biosynthesis, iron homeostasis and pathogen infection. We will thus engineer viral vectors to express or inhibit miR-122 and use them to dissect its role as key regulator of iron metabolism in livers of adult mice. Concurrently, we expect to gain novel insights into the ncRNA mechanisms underlying frequent iron-related disorders. In an ensuing phase, we will use our vectors to study the relevance of miR-122 and its downstream effects on iron metabolism for malaria whose causative agent (Plasmodium) depends on hepatic iron (collaboration with A.-K. Mueller, Heidelberg). As a whole, our project will significantly expand our understanding of the role of ncRNAs in liver metabolism and infection.